Article

RONIN Is an Essential Tran

scriptional Regulator ofGenes Required for Mitochondrial Function in theDeveloping Retina

Graphical Abstract

Highlights

d Ronin loss causes retinal progenitor cell-cycle arrest and

photoreceptor degeneration

d Striking phenocopy of Cyclin D1 nulls, but RONIN functions

independently

d RONIN is a regulator of mitochondrial gene expression

including ETC components

d RONIN coordinates mitochondrial activity and neural

progenitor proliferation

Poche et al., 2016, Cell Reports 14, 1684–1697February 23, 2016 ª2016 The Authorshttp://dx.doi.org/10.1016/j.celrep.2016.01.039

Authors

Ross A. Poche, Min Zhang,

Elda M. Rueda, ..., James F. Martin,

Thomas P. Zwaka, Mary E. Dickinson

Correspondencepoche@bcm.edu (R.A.P.),mdickins@bcm.edu (M.E.D.)

In Brief

Poche et al. identify Ronin (Thap11) as an

important regulator ofmitochondrial gene

expression that coordinates

mitochondrial activity and cell-cycle

progression. Loss of Ronin function leads

to specific deficits in the electron

transport chain as well as premature cell-

cycle exit, excessive neurogenesis, and

cell death.

Accession Numbers

GSE74830

Cell Reports

Article

RONIN Is an Essential TranscriptionalRegulator of Genes Required for MitochondrialFunction in the Developing RetinaRoss A. Poche,1,2,* Min Zhang,1 Elda M. Rueda,6 Xuefei Tong,1 Melissa L. McElwee,1 Leeyean Wong,1 Chih-Wei Hsu,1

Marion Dejosez,9,10 Alan R. Burns,6 Donald A. Fox,6,7,8 James F. Martin,1,2,3,5 Thomas P. Zwaka,9,10

and Mary E. Dickinson1,2,3,4,*1Department of Molecular Physiology and Biophysics2Program in Developmental Biology3Cardiovascular Research InstituteBaylor College of Medicine, Houston, TX 77030, USA4Department of Bioengineering, Rice University, Houston, TX 77005, USA5Texas Heart Institute, Houston, TX 77030, USA6College of Optometry7Department of Biology and Biochemistry8Department of Pharmacology

University of Houston, Houston, TX 77004, USA9Department of Developmental and Regenerative Biology10Black Family Stem Cell Institute

Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA

*Correspondence: poche@bcm.edu (R.A.P.), mdickins@bcm.edu (M.E.D.)http://dx.doi.org/10.1016/j.celrep.2016.01.039

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

SUMMARY

A fundamental principle governing organ size andfunction is the fine balance between cell proliferationand cell differentiation. Here, we identify RONIN(THAP11) as a key transcriptional regulator of retinalprogenitor cell (RPC) proliferation. RPC-specific lossof Ronin results in a phenotype strikingly similar tothat resulting from the G1- to S-phase arrest andphotoreceptor degeneration observed in the CyclinD1 null mutants. However, we determined that, ratherthan regulating canonical cell-cycle genes, RONINregulates a cohort of mitochondrial genes includingcomponents of the electron transport chain (ETC),which have been recently implicated as direct regula-tors of the cell cycle. Coincidentally, with prematurecell-cycle exit, Ronin mutants exhibited deficientETC activity, reduced ATP levels, and increasedoxidative stress that we ascribe to specific loss ofsubunits within complexes I, III, and IV. These dataimplicate RONIN as a positive regulator of mitochon-drial gene expression that coordinates mitochondrialactivity and cell-cycle progression.

INTRODUCTION

The mammalian CNS is composed of an astonishing variety

of morphologically and functionally distinct neuronal types

organized with precise regional specification and connectivity.

1684 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Auth

Throughout neurogenesis, cohorts of multi-potential neural pro-

genitor cells exit the cell cycle and differentiate, gradually

reducing the pool of proliferative progenitors over time. If too

many progenitor cells prematurely exit the cell cycle and termi-

nally differentiate, this depletes the progenitor pool for later

differentiating neurons, resulting in hypoplasia and degenera-

tion. Conversely, if neural progenitors fail to exit the cell cycle,

hyperplasia, dysplasia, and tumor formation can result. Despite

the strict requirement for coordinating progenitor proliferation

with cell-cycle exit and differentiation in CNS development, the

precise cellular and molecular mechanisms orchestrating these

events are poorly understood.

The mouse retina is relatively simple in structure with only

seven major cell types and is an excellent model system for

studying mammalian CNS development. Retinogenesis begins

at embryonic day 11 (E11.0) in a conserved, stereotypical order.

Retinal ganglion cells (RGCs) emerge first, followed by tempo-

rally overlapping phases of horizontal, cone, amacrine, rod,

bipolar, and M€uller glial cell genesis (Sernagor, 2006; Young,

1985). Since all seven cell types are all derived from a common

retinal progenitor cell (RPC) pool, the mouse retina is particularly

well suited for studying how proliferation and differentiation are

balanced so that all cell types are produced at the correct time

and ratios for proper cytoarchitecture and visual processing.

Despite advances in knowledge about transcriptional mecha-

nisms driving RPC fate specification and differentiation, the

manner in which RPC proliferation is regulated and coordinated

with intrinsic fate specifying factors remains obscure. This is

particularly true for transcriptional control of RPC proliferation.

Here, we examined the role of the embryonic stem cell

(ESC) pluripotency factor RONIN (THAP11) during mouse

ors

retinogenesis. RONIN was previously identified as a zinc-finger

transcriptional regulator that is essential for ESC self-renewal

and growth (Dejosez et al., 2008, 2010). Recently, several

in vitro studies have alluded to an additional role as a direct regu-

lator of cell proliferation (Parker et al., 2012, 2014). Therefore, we

hypothesized that RONIN might play a role in balancing the pro-

liferation of multipotent RPCs and neuronal differentiation. Using

a conditional knockout (CKO) approach, we found that RONIN is

indeed a key regulator of RPC proliferation. Specifically, CKO of

Ronin in RPCs results in a striking phenocopy of the Cyclin D1

null mutants in which RPCs undergo premature cell-cycle exit

leading to a thin, hypoplastic adult retina (Das et al., 2009; Sicin-

ski et al., 1995). Additionally, Ronin CKO retinae exhibit an un-

usual pattern of discontinuous photoreceptor degeneration

that was previously described as a unique feature of the Cyclin

D1 mutants (Ma et al., 1998).

Based on emerging evidence that CYCLIN D1 acts as a retinal

transcription factor (Bienvenu et al., 2010), we hypothesized that

CYCLIN D1 and RONIN might function cooperatively in a tran-

scription factor complex to directly control the expression of

genes essential for regulating RPC proliferation, but we deter-

mined that Ronin and Cyclin D1 do not interact at the genetic

or protein level. Also, in contrast to recent in vitro reports (Parker

et al., 2012, 2014), we did not identify enrichment of RONIN

target genes within canonical cell-cycle pathways. Rather, we

found that RONIN acts as a direct transcriptional regulator of nu-

clear-encoded mitochondrial genes. In particular and coincident

with premature cell-cycle exit, Ronin CKO retinae suffer from a

deficit in the electron transport chain (ETC) due to the de-regula-

tion of genes that encode subunits of complex I, III, and IV. We

also observed a reduction in ATP levels, oxidative stress, and

the activation of mitochondrial quality-control pathways. This

study identifies RONIN as a key transcription factor regulating

both progenitor cell proliferation and mitochondrial activity dur-

ing neural development.

RESULTS

Ronin Expression and RPC-Specific KnockoutqRT-PCR showed that Ronin transcripts were abundant in the

E14.5 retina, decreasing by 2-fold by postnatal day 50 (P50) (Fig-

ure 1A). Similarly, immunofluorescence data showed a clear RO-

NIN signal at E16.5 within the RPCs of the outer neuroblastic

layer (onbl) and inner neuroblastic layer (inbl) with fainter signals

within the ganglion cell layer (GCL) (Figure 1B). P20 retinae

showed ubiquitous RONIN protein expression (Figure 1B). Anal-

ysis of a Ronin-GFP knockin allele showed ubiquitous expres-

sion of GFP (Figure 1C). These data suggested that Ronin might

play a role in retinogenesis and possibly an additional role in

mature retinal neurons.

Since germline loss of Ronin leads to preimplantation lethality

(Dejosez et al., 2008), we generated Chx10-Cre::GFP+/tg;

Roninflox/flox CKO mice in which Ronin loss occurs specifically

within RPCs (Rowan and Cepko, 2004). CKO mice exhibited a

dramatic reduction of Ronin transcripts at E14.5 compared to

Roninflox/flox controls (Figure 1D). Immunofluorescence at E16.5

further showed that RONIN protein was also lost within the

CKO RPCs and their descendant neurons in the GCL (compare

Cell R

Figures 1B and 1E). We occasionally observed discrete patches

of RONIN expression that persisted within the RPC population

(arrow in Figure 1E) and visualization of CRE::GFP fluorescence

indicated that these RONIN+ patches correspond to regions

where Chx10-Cre::GFP is not expressed (arrowheads in Fig-

ure 1F). This finding suggests a low level of Chx10-Cre::GFP

mosaicism, consistent with previous reports (Rowan and

Cepko, 2004), and also confirms RONIN antibody specificity.

Despite these occasional patches of residual RONIN expression,

western blot analysis of Ronin CKO retinae confirmed that

there was a general and dramatic decrease in RONIN protein

beginning at E12.5 that became progressively more severe

(Figure 1G).

Ronin CKO Phenocopies the Adult Cyclin D1 NullMutantsRonin CKO mice are viable, and analysis of the adult retinae re-

vealed a striking structural phenotype. z stack confocal projec-

tions of DAPI-stained retinal flat mounts showed dramatic

‘‘holes’’ scattered throughout the photoreceptor layer in the

CKO retinae that were not observed in controls (Figure 2A).

This type of lesion was previously reported to be a unique feature

of the Cyclin D1 mutants, arising from localized photoreceptor

degeneration (Ma et al., 1998), and the similarity was confirmed

by direct comparison (Figure 2A). Both mutants had thin, hypo-

plastic retinae, suggesting a proliferation defect (Figure 2B), as

well as inner nuclear layer (INL) cells infiltrating the degenerated

outer nuclear layer (ONL) (Figure 2B, arrowheads). We also

observed a striking Cyclin D1 phenocopy at the level of

RHODOPSIN loss near the degenerative lesions (Figure 2C, ar-

rowheads) and correspondingGFAP+ reactive gliosis (Figure 2D)

indicating that both Ronin and Cyclin D1 mutants exhibit patchy

retinal degeneration. Thus, Ronin loss within the embryonic

retina impacts RPC proliferation and retinal neuron survival in a

similar manner as Cyclin D1 loss (Das et al., 2009; Sicinski

et al., 1995). Consistent with these findings, CKO of Ronin

broadly within the developing CNS recapitulated the ataxia and

cerebellar hypoplasia described for theCyclin D1/D2 doublemu-

tants (Ciemerych et al., 2002) (Figure S1; Movies S1 and S2).

Ronin Loss Results in Reduced RPC Proliferation andPremature Exit from the Cell CycleQuantification of the #EdU+/#DAPI+ pixels revealed a statisti-

cally significant decrease in S-phase entry as early as E14.5

and also at E18.5 (Figures 3A and 3B), that was also consistent

with a reduction in KI67 immunofluorescence (Figure 3C). As

another measure of Ronin CKO RPC proliferative capacity, we

performed replication incompetent retroviral lineage analysis

using either the NIN-E (control) or NIN-Cre (Cre-expressing) vi-

ruses, both of which encode a nuclear LacZ reporter and only

infect proliferative RPCs (Dyer, 2003; Dyer and Cepko, 2001).

Quantification of LacZ+ cells, examined 10 days following retinal

explant and infection at E14.5 (Figure 3D), revealed that the NIN-

Cre infected Roninflox/flox retinae contained a significant increase

in the percentage of single-cell clones and decrease in the per-

centage of clones that contained five or more cells as compared

to the NIN-E infected retinae (Figure 3E). These data indicated

that Ronin loss within RPCs directly contributes to premature

eports 14, 1684–1697, February 23, 2016 ª2016 The Authors 1685

Figure 1. Ronin Expression and Retinal-Specific Knockout

(A) qRT-PCR time course of Ronin expression in the wild-type retina. Levels are given as relative to P50 (set to 1) and depicted as fold changes ± SEM (n = 3

independent pooled samples per time point; Student’s t test).

(B) Ronin immunofluorescence in the E16.5 and P20 retina.

(C) Retinal GFP expression of Ronin+/GFP knockin allele.

(D) Reduction of RoninmRNA expression in E14.5 RPC-specific CKOs. Levels are given as relative to control (set to 1) and depicted as fold change ± SEM (n = 3

independent pooled samples per group; Student’s t test).

(E) Dramatic loss of Ronin protein within E16.5 RPCs. The arrow indicates an occasional patch of RPCs where Ronin expression persists.

(F) Ronin-positive patch corresponds to an area where Cre is not expressed.

(G) Western blot showing reduction of Ronin throughout retinogenesis.

cell-cycle exit in a cell-autonomous manner. However, we also

detected a small but significant increase in activated CAS-

PASE-3+ cells in the Ronin CKO retinae beginning at E16.5

(Figure S2A), further increasing by E18.5 and localized to the neu-

roblastic layer (NBL) and GCL (Figures S2A and S2B). Therefore,

we cannot rule out the possibility that Ronin loss also triggers

apoptosis that contributes to changes inmutant clone size. Addi-

tional analysis of apoptosis, using the ROSA26R-mTmG+/tg Cre

reporter (Muzumdar et al., 2007) and RPC-specificmarkers, sug-

gested that Ronin loss likely affects RPC and neuron survival in

what is likely to be a cell-autonomous manner (Figure S2).

1686 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Auth

We also examined cell proliferation in Ronin retinal CKOs us-

ing the Fucci cell-cycle reporter mice, which label cells in the

G1 phase with a red fluorescent protein and a GFP once cells

enter the S phase and progress to G2 and M phase (Sakaue-

Sawano et al., 2008). Post-mitotic retinal neurons also maintain

red fluorescence (Sakaue-Sawano et al., 2008). We found that

P0 Ronin CKO RPCs exhibited dramatically fewer cells in S,

G2, and M phases, while the red fluorescence within the GCL

appeared increased (Figure 3F). These results suggested that

there are fewer proliferating RPCs in the Ronin CKOs and

that premature exit from the cell cycle leads to overproduction

ors

Figure 2. Ronin CKOs Phenocopy the

Cyclin D1�/� Retina

(A) Retinal flat mounts stained with DAPI high-

lighting the holes scattered through out the Ronin

and Cyclin D1 mutants.

(B) Cryosections showing cross sections of the

retinal holes (arrowheads) and thinning of the

mutant retinae.

(C) Rhodopsin immunofluorescence showing loss

within the mutant retinal lesions (arrowheads).

(D) GFAP immunofluorescence showing reactive

gliosis (arrow) in themutant retinae with astrocytes

projecting into the retinal holes (arrowheads).

See also Figures S1 and S3.

of neurons. Consistent with this interpretation, E18.5 DAPI-

stained sections showed that the RPC-containing NBL layer

of the developing Ronin CKO retina was thinner than controls,

while the post-mitotic, neuronal GCL appeared to be expanded

(Figure 3G, green bars and red bars). We also observed a

decrease in immunofluorescent labeling for the RPC-expressed

transcription factor SOX2 (Figure 3G, white bars) (Taranova

et al., 2006).

Cell Reports 14, 1684–1697, F

If the decrease in the Ronin mutant

RPC population resulted from prema-

ture cell-cycle exit, as in the Cyclin D1

mutants, we would expect to observe

a consequential increase in the early

retinal neuron population (Das et al.,

2009). Using q-RT-PCR, we showed

that the pro-neural gene NeuroD1 was

upregulated in both the Ronin CKO

and Cyclin D1 null E15.5 retinae, sug-

gesting that both mutants exhibit exces-

sive neurogenesis at the expense of the

RPC pool (Figure 3H). NEUROD1 drives

mouse retinogenesis and is transcrip-

tionally upregulated in response to

both Notch1 and Cyclin D1 loss (Hata-

keyama and Kageyama, 2004; Yaron

et al., 2006). Additionally, the domain

of expression of pan-neuronal marker

TUJ1 was also expanded in the Ronin

CKO retinae at P0, coinciding with re-

gions of CRE activity, again suggesting

a cell-autonomous effect of Ronin loss

(Figure 3I, arrows). Finally, we labeled

E18.5 retinal flat mounts with anti-

bodies against the RGC-specific marker

BRN3B and imaged the GCL en face

with confocal microscopy. We observed

an increase in BRN3B+ RGCs within

the E18.5 Ronin CKO retinae (Figures

3J and 3K). In total, these data support

the conclusion that Ronin loss leads

to premature exit of RPCs from the

cell cycle resulting in excessive

neuronal differentiation in a similar fashion to the Cyclin D1

mutant retina.

RONIN Is a Direct Transcriptional Regulator ofMitochondrial Gene Expression, Including GenesRequired to Form ETC ComplexesDue to the similarities between the Ronin and Cyclin D1 mutant

retinae, and the recently established role for CYCLIN D1 as a

ebruary 23, 2016 ª2016 The Authors 1687

retinal transcription factor (Bienvenu et al., 2010), we reasoned

that RONIN and CYCLIN D1 might functionally cooperate in a

protein complex to regulate a specific cohort of genes that

influence RPC proliferation. However, to our surprise, we found

that RONIN and CYCLIN D1 do not interact genetically or at

the protein level (Figure S3). Therefore, we hypothesized that

RONIN-specific transcriptional targets are likely responsible for

RPC cell-cycle arrest.

To identify these genes, we performed RONIN chromatin

immunoprecipitation from P0 wild-type retinae followed by

DNA sequencing (ChIP-seq). We found 904 RONIN-bound re-

gions enriched within promoter regions throughout the genome,

consistent with a role for RONIN as a transcriptional regulator

(Figure 4A). Of these 904 RONIN ChIP-seq peaks, covering

898 genes, 305 also contained a conserved Ronin binding motif,

whichwas previously identified in ESCs, andwas enrichedwithin

promoter regions (Figure 4B). To focus our attention on the phys-

iologically relevant ChIP-seq targets that are both expressed

in RPCs and affected by Ronin loss, we performed RNA

sequencing (RNA-seq) on E14.5 Ronin CKO and control retinae.

This experiment revealed 124 genes that were upregulated and

247 that were downregulated in response to Ronin loss (Fig-

ure S4; Table S1). Overlay of these data with the ChIP-seq data-

set revealed that 26 out of 124 upregulated genes and 99 out of

247 downregulated genes are bound by RONIN (Figures 4C and

4D; Tables S1 and S2), suggesting that RONIN functions as both

a transcriptional repressor and activator during retinogenesis.

Gene Ontology (GO) analysis of the overlapping ChIP-seq and

RNA-seq data showed that the upregulated genes are involved

in proteolysis, peptidase activity, and oxidation-reduction pro-

cesses (Figure 4E; Table S3). Among the downregulated genes,

we found enrichment for Ras GTPase binding, zinc ion binding,

and mitochondrial genes (Figure 4F; Table S3). To our surprise,

we did not detect significant enrichment of genes implicated in

direct cell-cycle regulation or neurogenesis. Based on mounting

evidence of the connection between mitochondrial activity and

cell-cycle regulation (Lee and Finkel, 2013; Mandal et al., 2005;

Mitra et al., 2009; Owusu-Ansah et al., 2008; Schieke et al.,

2008; Wang et al., 2014), we chose to investigate the RONIN

mitochondrial target genes in more detail.

Figure 5A shows a heatmap derived from the RNA-seq targets

associated with the GO term ‘‘mitochondrion’’ that exhibited

downregulation in the Ronin CKO retinae. Selected examples

of the ChIP-seq peaks for a subset of these genes are shown

in Figure 5B illustrating RONIN binding sites localized to pro-

moter regions, and qRT-PCR analysis confirmed that these

genes are downregulated in the Ronin CKOs (Figure 5C). There-

fore, our ChIP-seq and RNA-seq screens identified genes ex-

pressed in the mitochondria of the developing retina that require

RONIN activity for transcriptional activation. Mapping of mito-

chondrial-annotated RONIN ChIP-seq peaks to a schematic of

a mitochondrion revealed that RONIN target genes are localized

to different mitochondrial compartments and have various pre-

dicted functions (Figure 5D). However, we noticed two distinct

clusters of genes with clearly defined roles in regulating ETC ac-

tivity. One cluster (light blue box in Figure 5D) encodes subunits

of ETC protein complexes I, III, IV, and V. The second cluster

(dark purple box in Figure 5D) encodes three mitochondrial ribo-

1688 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Auth

somal proteins (MRPL24, MRPL34, and MRPL54) and the

mitochondrial translational release factor (MTRF1). Since the

13 mitochondrial-encoded proteins are all core subunits of

ETC complexes, the disruption of mitochondrial translation

also negatively impacts ETC activity and results in a profound

mitochondrial functional deficiency.

Ronin CKO Retinae Show Signs of Oxidative Stress andMitochondrial DamageWe next determined whether Ronin loss in the retina indeed

results in compromised mitochondrial function and damage

coincident with RPC cell-cycle arrest. First, we assessed general

mitochondrial morphology within control and Ronin CKO P0

retinae by transmission electron microscopy (TEM). We noticed

that the CKO retinal onbl contained what appeared to be swollen

mitochondria with a loss of normal cristae morphology whereas

control mitochondria had pronounced cristae (Figure 6A).

However, TEM does not allow for the resolution of RPCs

versus immature neurons. Furthermore, due to the low level

Cre-mosaicism, we could not definitively identify Ronin CKO

versus wild-type cells. Therefore, we performed additional histo-

logical assays as confirmation of mitochondrial dysfunction.

Freshly isolated retinal whole mounts were stained with the

mitochondrial superoxide indicator MitoSOX, and we found

that Ronin CKO retinae exhibited increased labeling within the

onbl as compared to the negative controls (Figure 6B). By also

imaging Chx10-Cre::GFP expression within the Ronin CKOs,

we determined that the increase in MitoSOX labeling occurs in

GFP-negative cells, suggesting that mitochondrial superoxides

have accumulated predominantly within postmitotic cells versus

RPCs. Consistent with this finding, immunofluorescence for the

mitochondrial antioxidant manganese superoxide dismutase

(MNSOD), which is a often increased during periods of oxidative

stress (Dhar and St Clair, 2012), exhibited an increase within the

CKO onbl, INL, and GCL at P0 as compared to controls (Fig-

ure 6C), and this was confirmed by western blot (Figure 6D).

We next performed the OxyBlot assay and found that the Ronin

CKOprotein lysates had increased carbonylation, which is a hall-

mark of oxidative stress (Davies, 1987) (Figure 6E). Finally, we

also noticed that Parkin (Park2)was one of the most upregulated

transcripts in the Ronin CKO RNA-seq dataset, and qRT-PCR

showed that its increase is almost 5-fold (Figures 6F and 6G).

However, it is not enriched in our ChIP-seq data. Mutations in

PARK2 can cause Parkinson’s disease and encode a protein

that is essential for regulating mitochondrial quality control

(Dagda and Chu, 2009; Shimura et al., 2000). Thus, its transcrip-

tional upregulation is likely the consequence of mitochondrial

dysfunction caused by Ronin loss.

Ronin CKOs Exhibit a Specific Deficit in MitochondrialETC ActivityPrevious work in Drosophila indicated that genetic ablation of

specific ETC protein subunits drives a G1 to S phase cell-cycle

arrest (Mandal et al., 2005; Owusu-Ansah et al., 2008). Of the

RONIN ChIP-seq targets associated with the mitochondria,

seven encoded proteins that are localized to the ETC complexes

(Figure 7A) and all contained RONIN binding sites within their

promoters (Figure 7B). To determine if ETC protein complex

ors

Figure 3. Ronin Loss Results in Premature Cell-Cycle Exit and Excessive Neurogenesis

(A and B) EdU labeling of Ronin CKO retinae represented as the average ratio ± SEM of EdU-positive pixels to DAPI-positive pixels (n = 3 samples per group;

Student’s t test).

(C) CKOs show a reduction in immunofluorescence for Ki67.

(D) Representative images of nuclear LacZ-labeled control (NIN-E) and Ronin mutant (NIN-Cre) clones.

(legend continued on next page)

Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Authors 1689

Figure 4. ChIP-Seq and RNA-Seq Analyses

(A) Distribution of Ronin binding sites throughout

the P0 retinal genome and significant enrichment

at promoter regions.

(B) Enrichment of a known Ronin binding motif

within ChIP-seq peaks.

(C and D) Overlap of ChIP-seq with E14.5 Ronin

CKO RNA-seq data showing upregulation (C) and

downregulation (D).

(E and F) Gene Ontology (GO) analysis of the

genes upregulated and downregulated in the

ChIP-seq/RNA-seq overlap that showed a signif-

icant Z score. The numbers inside the bars

correspond to the number of genes within the

overlapping datasets that associate with that

particular GO term.

See also Figure S4.

formation requires Ronin, we performed qRT-PCR on Ronin

CKO and control retinae and found that Ndufa3, Uqcr10, and

Cox7c were transcriptionally downregulated. These genes

encode protein components of ETCcomplex I, III, and IV, respec-

tively. Interestingly, Ndufv1, Atp5d, and Atp5o showed slight

increases in mRNA levels, suggesting that they are possibly

negatively regulated by RONIN or upregulated via a compensa-

tory mechanism responding to an ETC deficiency (Figure 7C).

ETC complexes are large, multi-subunit complexes and have

been shown to exhibit reductions in stability and/or formation

(E) Scoring of clone composition is represented as the average percentage of clones ± SEM of a specific si

(F) Labeling of cell-cycle phases with the Fucci transgenes shows a decrease in cells within S/G2/M-phase.

(G) Ronin CKOs exhibit a diminishment of the Sox2+ RPC population.

(H) E15.5 Ronin and Cyclin D1mutants have increased NeuroD1 mRNA expression. Levels are given as relat

depicted as fold changes ± SEM (n = 3 independent pooled samples per group; Student’s t test).

(I) The domain of pan-neuronal marker Tuj1 expression is expanded in the Ronin CKOs.

(J and K) Ronin CKOs contain more Brn3b+ RGCs represented as the average ratio ± SEM of Brn3b-positiv

group; Student’s t test).

See also Figures S2 and S3.

1690 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Authors

when complex subunits are missing and

when partner complexes within ETC

supercomplexes are disrupted (Acin-

Perez and Enriquez, 2014; Acın-Perez

et al., 2008). We investigated whether

the reduction in Ndufa3, Uqcr10, and

Cox7c mRNA levels observed in the

Ronin CKO retinae impact the presence

of ETC protein complexes. Western blot

analysis showed both E15.5 and P0

Ronin CKO retinae exhibited a significant

reduction in NDUFA9, UQCRC2, and

COX4 proteins corresponding to com-

plex I, III, and IV, respectively. However,

complex II protein SDHA and complex V

protein ATP5A were unaffected (Fig-

ure 7D). We further confirmed this finding

by using an independent set of ETC anti-

bodies and found that NDUFB8,

UQCRC1, and COX1 proteins were all

dramatically reduced, again suggesting disruption of complex

I, III, and IV. SDHB and ATP5B levels were unchanged, providing

additional evidence that complex II and V are not affected (Fig-

ure 7E). Importantly, we did not detect any changes in levels of

non-ETC mitochondrial proteins VDAC1 or POLG (Figure 7F).

We also crossed the mito-Dendra2 mitochondrial fluorescent

transgene reporter into the Ronin CKOs, and we did not observe

any changes in mito-Dendra2 protein levels (Figure 7G) (Pham

et al., 2012). Taken together, these data suggest that Ronin

loss does not generally impact mitochondrial content. Rather,

ze (n = 4 samples per group; Student’s t test).

ive to independent littermate controls (set to 1) and

e pixels to DAPI-positive pixels (n = 3 samples per

Figure 5. Ronin Regulates a Diverse Cohort of Mitochondrial Genes

(A) Heatmap from Ronin CKO RNA-seq showing 30 mitochondrial genes that are downregulated (fold change R 1.5, FDR % 0.2).

(B) Images of Ronin ChIP-seq peaks within promoter regions shown for a subset of the genes in the RNA-seq heatmap. The red boxes denote the presence of a

Ronin binding motif.

(C) E15.5 qRT-PCR analysis of Ronin CKO retinae confirming a reduction in mRNA expression of representative ChIP-seq targets. Levels are given as relative to

the control (set to 1) and depicted as fold changes ± SEM (n = 3 independent pooled samples per group; Student’s t test).

(D) Genes with a Ronin ChIP-seq peak specifically within the promoter region that also show transcriptional downregulation either by RNA-seq, qRT-PCR or both

(red), upregulation by qRT-PCR (green), or no change by RNA-seq and qRT-PCR (black) were mapped to a schematic of a mitochondrion and functionally

annotated.

Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Authors 1691

the mRNA reduction we observed at the level of the Ronin target

genes Ndufa3, Uqcrc10, and Cox7c likely impacts the stability

and/or formation of ETC complexes I, III, and IV.

As a functional assessment of ETC, we performed cytochrome

c oxidase (COX) histochemistry on E15.5 and P0 retinal cryosec-

tions. Despite the findings that the mito-Dendra2 transgene

clearly showed mitochondria present in neurons within the P0

Ronin CKOs GCL and inner plexiform layer (IPL) (Figure 7H, ar-

row) and that the mtDNA content was actually increased (Fig-

ure 7I), we uncovered a dramatic reduction in COX activity, which

was obvious as early as E15.5 and became progressively more

severe by P0 (compare Figures 7J–JL). This result shows that a

transcriptional reduction in key ETC RONIN target genes ulti-

mately results in a deficiency in complex IV activity in vivo.

Consistent with the ETC deficit, we showed that P0 Ronin CKO

retinae exhibited a significant (almost 50%) reduction in ATP

levels relative to controls (Figure 7M). Thus, taken together,

these data strongly implicate Ronin as a major regulator of

ETC activity in the developing retina.

DISCUSSION

We have uncovered a role for RONIN (THAP11) as a transcrip-

tional regulator that profoundly influences RPC proliferation

and retinal neuron survival. By conditionally knocking out Ronin,

we showed that the balance of RPC proliferation and differenti-

ation is disrupted as RPCs prematurely exit the cell cycle and

the retina experiences excessive embryonic neurogenesis and

apoptosis. Ultimately, this results in a thin, underdeveloped adult

tissue that also phenocopies the discontinuous pattern of photo-

receptor degeneration previously described as a unique feature

of the Cyclin D1mutant mice (Ma et al., 1998). However, molec-

ular and genetic interaction experiments revealed that RONIN

functions independently of CYCLIN D1. To our surprise, ChIP-

seq and RNA-seq experiments identified a cohort of nuclear-en-

coded mitochondrial genes that are bound by RONIN and are

significantly reduced inRoninCKO retinae. Seven of these target

genes encode ETC complex subunits and are directly required

for ETC function, which has been directly implicated in cell-cycle

G1 to S phase regulation (Mandal et al., 2005; Owusu-Ansah

et al., 2008). Our data indicate that dysregulation of mitochon-

drial function following Ronin loss has a significant impact on

the normal proliferation, differentiation, and survival of retinal

progenitor cells.

Our studies also revealed a large group of zinc ion binding

genes enriched in our Ronin retinal ChIP-seq and RNA-seq

datasets. However, none of these factors have been clearly

implicated in cell-cycle regulation, mitochondrial activity, or ret-

inogenesis. Future efforts will be aimed at clarifying their roles in

the retina.

RONIN as a Major Regulator of Mitochondrial GeneExpression and ActivityIn our analyses of the Ronin CKO retinae, we observed distinct

signs ofmitochondrial oxidative stress and the activation of qual-

ity-control pathways, which we attribute to loss of nuclear-en-

coded RONIN target genes necessary for proper mitochondrial

function. Our finding that RONIN plays such a profound role in

1692 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Auth

regulating the expression of proteins required for mitochondrial

biogenesis and function is significant, because only a handful

of such nuclear-encoded transcription factors have been identi-

fied, including PGC-1a, NRF-1, NRF-2, and TFAM (Scarpulla

et al., 2012; Wu et al., 1999). Therefore, our data raise the

obvious question of whether RONIN functionally interacts with

these factors. Interestingly, the zebrafish mutant for not really

finished (nrf), the homolog of mammalian Nrf-1, was shown to

exhibit a thin, underdeveloped retina with patchy, degenerative

lesions reminiscent of the Ronin and Cyclin D1 mutants (Becker

et al., 1998). In Drosophila, Cyclin D is a positive regulator of

mitochondrial function by influencing the activity of nrf-1 and

its target gene, tfam (Icreverzi et al., 2012; Icreverzi et al.,

2015). However, there are also contradictory data suggesting

that mammalian CYCLIN D1 is actually a negative regulator of

NRF-1 activity, inhibiting mitochondrial biogenesis (Sakamaki

et al., 2006; Wang et al., 2006). Based on the co-immunoprecip-

itation results presented here, we do not believe that there is a

direct protein/protein interaction between RONIN and CYCLIN

D1, and we have P0 retinal immunoprecipitation data suggesting

that interactions between NRF-1 and RONIN or its co-factor,

HCF1, do not occur (Figure S5). Thus, our understanding of the

interactions between mitochondrial transcription factors is far

from complete, and future studies should be focused on identi-

fying additional RONIN co-factors.

A Role for RONIN in Regulating RPC Proliferation andRetinal Neuron SurvivalRONIN has been identified as a transcriptional regulator that

influences ESC pluripotency and growth, but a role in regulating

proliferation and/or differentiation during neural development

in vivo has not been described (Dejosez et al., 2008, 2010).

However, a recent in vitro study has suggested an additional

role for RONIN in transcriptional regulation of the cell cycle

(Parker et al., 2014). HeLa cells were shown to contain a

RONIN/HCF-1/ZNF143 transcription factor complex that is

enriched on E2F-bound promoters of genes implicated in cell

proliferation. While these genes are represented in our RONIN

ChIP-seq dataset and we cannot conclusively rule them out as

contributing to the Ronin CKO retinal phenotype, only one of

these genes (Bod1) was significantly reduced within our RNA-

seq dataset. Furthermore, we have not been able to detect the

presence of ZNF143 protein in RONIN and HCF1 co-immuno-

precipitations from P0 mouse retinal lysates (data not shown).

It is also important to note that while the E2f1/2/3 triple-mutant

retinae have increased RPC apoptosis, unlike Ronin CKOs, it is

specifically restricted to the P0 time point. Additionally, these

mice have a modest proliferation defect, and there is no report

of the discontinuous pattern of photoreceptor degeneration

observed in the Ronin and Cyclin D1mutants (Chen et al., 2009).

Based on our finding that RONIN is an important transcrip-

tional regulator ETC components and othermitochondrial genes,

we propose that RONIN impacts RPC proliferation and retinal

neuron survival at the level of mitochondrial activity rather than

cell-cycle gene expression. This idea is consistent with several

groundbreaking studies in Drosophila that implicated ETC com-

plex I and III in directly influencing the G1 to S phase transition

(Mandal et al., 2005; Owusu-Ansah et al., 2008). Specifically,

ors

Figure 6. Ronin CKO Retinae Exhibit Signs of Oxidative Stress

(A) TEM image of control and Ronin CKO P0 retinae.

(B) MitoSOX Red staining and Cre::GFP expression in P0 Roninflox/flox (negative control), Ronin CKOs, and Roninflox/flox treated with Antimycin A (positive control)

retinal whole mounts imaged at the level of the onbl.

(C) MnSOD immunofluorescence indicated an increase within the P0 Ronin CKO INL and GCL.

(legend continued on next page)

Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Authors 1693

we found thatRoninCKO retinae showed a reduction in ETC pro-

tein complex I, III, and IV assembly and/or stability coincident

with premature exit from the RPC cell cycle. We attribute this

to a decrease in expression of RONIN target genes Ndufa3,

Uqcr10, and Cox7c, whose gene products reside in these com-

plexes. It is especially interesting that we observed reductions in

ETC complex I, III, and IV, as these are known to form a function-

ally interdependent supercomplex called the respirasome. Upon

disruption of one of these complexes, the partners within the

supercomplex are affected and overall activity is compromised

(Acın-Perez et al., 2008; Moreno-Lastres et al., 2012; Schafer

et al., 2006). This is also reflected in our finding that Ronin

CKO retinal COX activity and ATP levels are reduced.

RONIN and CYCLIN D1 Function Likely Converge uponMitochondrial ActivityGiven the remarkable phenocopy between Ronin and Cyclin D1

mutant retinae and cerebella, we were surprised that we did not

detect either a genetic or molecular interaction. CYCLIN D1

recently emerged as a protein of immense functional diversity

with roles not only within the canonical cell-cycle pathway but

also in transcriptional regulation, DNA repair, glucose meta-

bolism, and apoptosis (Bienvenu et al., 2010; Choi et al., 2014;

Jirawatnotai et al., 2011, 2012; Lee et al., 2014). Therefore, there

may be aspects of CYCLIN D1 function, possibly relating to

RONIN, that remain to be discovered. We favor the hypothesis

that RONIN and CYCLIN D1 play independent roles and the

commonality they share is at the level of regulatingmitochondrial

activity. Furthermore, the unusual pattern of photoreceptor

degeneration observed in both the Ronin CKO and the Cyclin

D1 null retinae (Ma et al., 1998) might result from aberrant mito-

chondrial function.

In the case of RONIN, we presented integrative and multi-

disciplinary results that support its role as a transcriptional

regulator of mitochondrial gene expression that, when lost,

compromises the ETC and likely other mitochondrial functions.

It is possible that CYCLIN D1 might also regulate mitochondrial

gene expression, albeit via a separate cohort from those

regulated by RONIN. It is also conceivable that CYCLIND1might

influence mitochondrial function at the protein level. Recent

CYCLIN D1 mass spectrometry data suggest possible interac-

tions with mitochondrial proteins (Bienvenu et al., 2010). Even

more tantalizing is a recent report that CYCLIN B1 and associ-

ated CDKs physically transport to the mitochondria to phos-

phorylate ETC components, which increases ATP production

in preparation for M phase (Wang et al., 2014). CYCLIN D1might

play a similar role as CYCLIN B1, but at the G1 to S phase

transition. While there is still much to be learned about the role

mitochondria play in directly regulating the cell cycle, our study

identifies RONIN as a transcriptional regulator of mitochondrial

gene expression with significant impact on future investiga-

tions of the transcriptional networks coordinating mitochondrial

(D) Western blot of E15.5 and E18.5 Ronin CKO confirmed a slight increase in M

(E) Independent Oxyblot assays of P0 retinal lysates showed that Ronin CKO ret

(F) Heat map from the Ronin CKO RNA-seq showing the increase in Park2 (fold c

(G) qRT-PCR confirmation that the Park2 mRNA level is increased in the Ronin C

change ± SEM (n = 3 independent pooled samples per group; Student’s t test).

1694 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Auth

activity with neural progenitor cell proliferation and terminal

differentiation.

EXPERIMENTAL PROCEDURES

See Supplemental Experimental Procedures for detailed protocols, lists of

reagents, statistics, and associated references.

Mouse Strains

The following mouse lines were used and PCR genotyped using published

protocols: Roninflox/flox, Ronin+/GFP, Chx10Cre::GFP+/tg, Fucci, Gfap-Cre+/tg,

ROSA26R-mTmG+/tg, Cyclin D1�/�, p27Kip1�/�, and Mito-Dendra+/tg. All

animal research was conducted according to protocols approved by the Insti-

tutional Animal Care and Use Committee of Baylor College of Medicine.

Histology and Microscopy

Immunofluorescence, X-gal staining, and confocal microscopy of cryosec-

tions were performed as previously described withminor modifications (Poche

et al., 2007). COX histochemistry was performed as previously described

(Wong-Riley, 1979). To label retinal progenitors in S phase, 5-ethynyl-20-deox-yuridine (EdU) was injected intraperitoneally (IP) into pregnant dams at a

concentration of 50 mg/kg body weight. After 2 hr, mice were sacrificed to

harvest embryos for retinal cryosections and EdU detection with the Click-iT

Plus EdU Alexa Fluor 555 Kit (ThermoFisher). For TEM, posterior eyecups

were fixed in cold 3%paraformaldehyde + 3%glutaraldehyde in 0.1M sodium

cacodylate for 2 days, post-fixed in 1% OsO4, dehydrated in an EtOH series,

and embedded in Epon. 80- to 90-nm sections were cut and stained with ura-

nyl acetate and lead citrate. To detect mitochondrial superoxide, freshly

dissected retinae were incubated with 5 mm MitoSOX Red (Molecular Probes,

#M36008), diluted in 13 PBS, for 10 min at 37�C followed by gentle washing in

13 PBS at 37�C. As a positive control, Roninflox/flox retinae were treated with

5 mM of the ETC complex III inhibitor antimycin A for 1 hr at 37�C prior to

staining.

RNA Isolation, qRT-PCR, and RNA-Seq

Retinae were dissected and total RNA immediately purified using the RNeasy

Mini Kit (QIAGEN) followed by reverse transcription using the SuperScript III

first-strand synthesis kit (Invitrogen). The TaqMan gene expression assay

from Applied Biosystems was used to assess differences between control

andmutant retinae. For RNA-seq, total RNA was poly-A selected and fraction-

ated (200–500 nt range). cDNA was generated using the Invitrogen Super-

Script III reverse transcription kit followed by second-strand cDNA synthesis

and then subjected to standard Illumina adaptor ligation and sequencing.

Alignment of RNA-seq data were performed using the bwa (version 0.7.6a)

analysis pipeline against mouse genome mm9. Significant differences were

called using cutoffs p value % 0.05, fold change R 1.5, and false discovery

rate (FDR) % 0.10.

Protein Isolation, Western Blot, Oxyblot, and Immunoprecipitation

Retinae were homogenized in Pierce lysis buffer (Thermo Fisher Scientific,

#87787) supplemented with Halt protease inhibitor cocktail (Thermo Fisher,

#78430). Protein was then run on 4%–15% Tris-HCl precast gels (Bio-Rad,

#3450027) and transferred (30 V, overnight) onto Immobilon-P PVDF Mem-

branes (Millipore) using the Criterion system (Bio-Rad). Blots from at least 3

independent controls andRoninCKOswere incubatedwith primary antibodies

(in 5% milk, overnight at 4�C) followed by incubation with horseradish-

peroxidase-conjugated secondary antibodies, detection with ECL Substrate

(Thermo Fisher, #32106), and exposure to film (Phenix, F-BX57). The OxyBlot

protein oxidation detection kit (EMDMillipore, #S7150) was used following the

nSOD protein at E15.5 and a significant increase by E18.5.

inae undergo excessive protein oxidation.

hange R 1.5, FDR % 0.2).

KOs. Levels are given as relative to the control (set to 1) and depicted as fold

ors

(legend on next page)

Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Authors 1695

manufacturer’s instructions. For details on co-immunoprecipitation experi-

ments, please see Supplemental Experimental Procedures.

ChIP, ChIP-qPCR, and ChIP-Seq

Chromatin immunoprecipitation was performed as described previously (Mao

et al., 2013). For ChIP-qPCR, to assess whether RONIN binds intron 2 of

Notch1, purified RONIN-bound DNA was amplified using SYBR Green PCR

master mix (ThermoFisher, #4309155). ChIP libraries were created and

sequenced using the Ion Torrent PGM system. Reads were mapped to the

mm9 assembly (NCBI Build 37) using Torrent Suite (2.0.1) aligner Tmap

(0.2.3) (Life Technologies). The cutoffs for calling the peaks were read-number

enriched at least 4-fold, with an FDR threshold of 0.001, FDR effective Poisson

p value of 1.3e-6, and minimum read number of 6.

ACCESSION NUMBERS

The accession number for the RNA-seq and ChIP-seq data (Edgar et al., 2002)

reported in this paper is GEO: GSE74830 (http://www.ncbi.nlm.nih.gov/geo/

query/acc.cgi?acc=GSE74830).

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures,

five figures, three tables, and twomovies and can be found with this article on-

line at http://dx.doi.org/10.1016/j.celrep.2016.01.039.

AUTHOR CONTRIBUTIONS

Conceptualization, R.A.P., and M.E.D.; Methodology, R.A.P., M.E.D., T.P.Z.,

M.D., and J.F.M.; Formal Analysis, R.A.P., M.Z., and J.F.M.; Investigation,

R.A.P., M.Z., E.M.R., X.T., M.L.M., L.W., and C.W.; Writing – Original Draft,

R.A.P.; Writing – Reviews and Editing, R.A.P. and M.E.D.; Resources,

A.R.B., M.D., D.A.F., J.F.M., and T.P.Z; Funding Acquisition, R.A.P., M.E.D.,

and J.F.M.

ACKNOWLEDGMENTS

We thank E. Levine for providing the Cyclin D1 null mice and M. Dyer for NIN-E

and NIN-Cre plasmids. We thank the Vector Development Lab (BCM) for pur-

ifying the NIN-E and NIN-Cre retroviruses and the Genomics Core Facility at

the Ichan School ofMedicine for assistancewith RNA-seq.We thankB. Reese,

J. Jankowsky, A. Groves, J. Culver, and M. Dyer for critical reading of the

manuscript. This work was supported by grants from the NIH (R01

EY024906 to R.A.P.; R01 EB016629 and P20 EB007076 to M.E.D.; R01

DE023177 and R01 HL118761 to J.F.M.; and P30 EY002520 and P30

EY007551), the Kyle and Josephine Morrow Endowed Chair (M.E.D.), the

Vivian L. Smith Foundation (J.F.M.), and the Knights Templar Eye Foundation

Career Starter Grant (R.A.P.).

Figure 7. Ronin Loss Results in a Functional Deficit in the ETC

(A) Schematic of the ETC highlighting genes that contain Ronin ChIP-seq peaks.

(B) ChIP-seq peaks within ETC genes. The red boxes denote the presence of a R

(C) qRT-PCR analysis of ETC target genes showing changes in mRNA levels. Lev

SEM (n = 3 independent pooled samples per group; Student’s t test). ***p < 0.00

(D) Western blot of protein subunits within the ETC showing a decrease in comp

(E) Independent set of antibodies confirming the finding in (D).

(F and G) Levels of the non-ETC mitochondrial proteins Vdac1 and Polg (F) and th

are unaffected by Ronin loss.

(H) Ronin CKO and control mice expressing mito-Dendra show that the Ronin C

(I) The Ronin CKOmtDNA content is increased relative to controls. Levels are give

samples per group; Student’s t test).

(J–L) Cytochrome c oxidase (COX) histochemistry of E15.5 and P0 Ronin CKOs

optical density values ± SEM (n > 3 independent samples per group; Student’s t

(M) Ronin CKOs have reduced ATP levels represented as the average micromol

Student’s t test).

1696 Cell Reports 14, 1684–1697, February 23, 2016 ª2016 The Auth

Received: November 12, 2015

Revised: December 14, 2015

Accepted: January 10, 2016

Published: February 11, 2016

REFERENCES

Acin-Perez, R., and Enriquez, J.A. (2014). The function of the respiratory super-

complexes: the plasticity model. Biochim. Biophys. Acta 1837, 444–450.

Acın-Perez, R., Fernandez-Silva, P., Peleato, M.L., Perez-Martos, A., and

Enriquez, J.A. (2008). Respiratory active mitochondrial supercomplexes.

Mol. Cell 32, 529–539.

Becker, T.S., Burgess, S.M., Amsterdam, A.H., Allende, M.L., and Hopkins, N.

(1998). not really finished is crucial for development of the zebrafish outer

retina and encodes a transcription factor highly homologous to human Nuclear

Respiratory Factor-1 and avian Initiation Binding Repressor. Development

125, 4369–4378.

Bienvenu, F., Jirawatnotai, S., Elias, J.E., Meyer, C.A., Mizeracka, K., Marson,

A., Frampton, G.M., Cole, M.F., Odom, D.T., Odajima, J., et al. (2010). Tran-

scriptional role of cyclin D1 in development revealed by a genetic-proteomic

screen. Nature 463, 374–378.

Chen, D., Pacal, M., Wenzel, P., Knoepfler, P.S., Leone, G., and Bremner, R.

(2009). Division and apoptosis of E2f-deficient retinal progenitors. Nature

462, 925–929.

Choi, Y.J., Saez, B., Anders, L., Hydbring, P., Stefano, J., Bacon, N.A., Cook,

C., Kalaszczynska, I., Signoretti, S., Young, R.A., et al. (2014). D-cyclins

repress apoptosis in hematopoietic cells by controlling death receptor Fas

and its ligand FasL. Dev. Cell 30, 255–267.

Ciemerych, M.A., Kenney, A.M., Sicinska, E., Kalaszczynska, I., Bronson, R.T.,

Rowitch, D.H., Gardner, H., and Sicinski, P. (2002). Development of mice

expressing a single D-type cyclin. Genes Dev. 16, 3277–3289.

Dagda, R.K., and Chu, C.T. (2009). Mitochondrial quality control: insights on

how Parkinson’s disease related genes PINK1, parkin, and Omi/HtrA2 interact

to maintain mitochondrial homeostasis. J. Bioenerg. Biomembr. 41, 473–479.

Das, G., Choi, Y., Sicinski, P., and Levine, E.M. (2009). Cyclin D1 fine-tunes the

neurogenic output of embryonic retinal progenitor cells. Neural Dev. 4, 15.

Davies, K.J. (1987). Protein damage and degradation by oxygen radicals. I.

general aspects. J. Biol. Chem. 262, 9895–9901.

Dejosez, M., Krumenacker, J.S., Zitur, L.J., Passeri, M., Chu, L.F., Songyang,

Z., Thomson, J.A., and Zwaka, T.P. (2008). Ronin is essential for embryogen-

esis and the pluripotency of mouse embryonic stem cells. Cell 133, 1162–

1174.

Dejosez, M., Levine, S.S., Frampton, G.M., Whyte, W.A., Stratton, S.A., Bar-

ton, M.C., Gunaratne, P.H., Young, R.A., and Zwaka, T.P. (2010). Ronin/

Hcf-1 binds to a hyperconserved enhancer element and regulates genes

involved in the growth of embryonic stem cells. Genes Dev. 24, 1479–1484.

onin binding motif.

els are given as relative to the control (set to 1) and depicted as fold changes ±

05 and *p < 0.01.

lex I, III, and IV in the CKOs.

e mitochondrial-targeted Dendra protein (from the mito-Dendra transgene) (G)

KO mitochondrial content is not reduced (arrow).

n as the average number of copies of mtDNA ± SEM (n = 3 independent pooled

revealed a dramatic decrease in COX activity. Levels were given as average

test).

es ATP per wet weight ± SEM (n = 3 independent pooled samples per group;

ors

Dhar, S.K., and St Clair, D.K. (2012). Manganese superoxide dismutase

regulation and cancer. Free Radic. Biol. Med. 52, 2209–2222.

Dyer, M.A. (2003). Regulation of proliferation, cell fate specification and differ-

entiation by the homeodomain proteins Prox1, Six3, and Chx10 in the devel-

oping retina. Cell Cycle 2, 350–357.

Dyer, M.A., and Cepko, C.L. (2001). p27Kip1 and p57Kip2 regulate prolifera-

tion in distinct retinal progenitor cell populations. J. Neurosci. 21, 4259–4271.

Edgar, R., Domrachev, M., and Lash, A.E. (2002). Gene Expression Omnibus:

NCBI gene expression and hybridization array data repository. Nucleic Acids

Res. 30, 207–210.

Hatakeyama, J., and Kageyama, R. (2004). Retinal cell fate determination and

bHLH factors. Semin. Cell Dev. Biol. 15, 83–89.

Icreverzi, A., de la Cruz, A.F., Van Voorhies, W.A., and Edgar, B.A. (2012).

Drosophila cyclin D/Cdk4 regulates mitochondrial biogenesis and aging and

sensitizes animals to hypoxic stress. Cell Cycle 11, 554–568.

Icreverzi, A., de la Cruz, A.F., Walker, D.W., and Edgar, B.A. (2015). Changes in

neuronal CycD/Cdk4 activity affect aging, neurodegeneration, and oxidative

stress. Aging Cell 14, 896–906.

Jirawatnotai, S., Hu, Y., Michowski, W., Elias, J.E., Becks, L., Bienvenu, F.,

Zagozdzon, A., Goswami, T., Wang, Y.E., Clark, A.B., et al. (2011). A function

for cyclin D1 in DNA repair uncovered by protein interactome analyses in

human cancers. Nature 474, 230–234.

Jirawatnotai, S., Hu, Y., Livingston, D.M., and Sicinski, P. (2012). Proteomic

identification of a direct role for cyclin d1 in DNA damage repair. Cancer

Res. 72, 4289–4293.

Lee, I.H., and Finkel, T. (2013). Metabolic regulation of the cell cycle. Curr.

Opin. Cell Biol. 25, 724–729.

Lee, Y., Dominy, J.E., Choi, Y.J., Jurczak, M., Tolliday, N., Camporez, J.P.,

Chim, H., Lim, J.H., Ruan, H.B., Yang, X., et al. (2014). Cyclin D1-Cdk4 controls

glucose metabolism independently of cell cycle progression. Nature 510,

547–551.

Ma, C., Papermaster, D., and Cepko, C.L. (1998). A unique pattern of photore-

ceptor degeneration in cyclin D1 mutant mice. Proc. Natl. Acad. Sci. USA 95,

9938–9943.

Mandal, S., Guptan, P., Owusu-Ansah, E., and Banerjee, U. (2005). Mitochon-

drial regulation of cell cycle progression during development as revealed by

the tenured mutation in Drosophila. Dev. Cell 9, 843–854.

Mao, C.A., Cho, J.H., Wang, J., Gao, Z., Pan, P., Tsai, W.W., Frishman, L.J.,

and Klein, W.H. (2013). Reprogramming amacrine and photoreceptor progen-

itors into retinal ganglion cells by replacing Neurod1 with Atoh7. Development

140, 541–551.

Mitra, K., Wunder, C., Roysam, B., Lin, G., and Lippincott-Schwartz, J. (2009).

A hyperfused mitochondrial state achieved at G1-S regulates cyclin E buildup

and entry into S phase. Proc. Natl. Acad. Sci. USA 106, 11960–11965.

Moreno-Lastres, D., Fontanesi, F., Garcıa-Consuegra, I., Martın, M.A., Arenas,

J., Barrientos, A., and Ugalde, C. (2012). Mitochondrial complex I plays an

essential role in human respirasome assembly. Cell Metab. 15, 324–335.

Muzumdar, M.D., Tasic, B., Miyamichi, K., Li, L., and Luo, L. (2007). A global

double-fluorescent Cre reporter mouse. Genesis 45, 593–605.

Owusu-Ansah, E., Yavari, A., Mandal, S., and Banerjee, U. (2008). Distinct

mitochondrial retrograde signals control the G1-S cell cycle checkpoint. Nat.

Genet. 40, 356–361.

Parker, J.B., Palchaudhuri, S., Yin, H., Wei, J., and Chakravarti, D. (2012). A

transcriptional regulatory role of the THAP11-HCF-1 complex in colon cancer

cell function. Mol. Cell. Biol. 32, 1654–1670.

Parker, J.B., Yin, H., Vinckevicius, A., and Chakravarti, D. (2014). Host cell

factor-1 recruitment to E2F-bound and cell-cycle-control genes is mediated

by THAP11 and ZNF143. Cell Rep. 9, 967–982.

Cell R

Pham, A.H., McCaffery, J.M., and Chan, D.C. (2012). Mouse lines with photo-

activatable mitochondria to study mitochondrial dynamics. Genesis 50,

833–843.

Poche, R.A., Kwan, K.M., Raven, M.A., Furuta, Y., Reese, B.E., and Behringer,

R.R. (2007). Lim1 is essential for the correct laminar positioning of retinal

horizontal cells. J. Neurosci. 27, 14099–14107.

Rowan, S., and Cepko, C.L. (2004). Genetic analysis of the homeodomain

transcription factor Chx10 in the retina using a novel multifunctional BAC

transgenic mouse reporter. Dev. Biol. 271, 388–402.

Sakamaki, T., Casimiro, M.C., Ju, X., Quong, A.A., Katiyar, S., Liu, M., Jiao, X.,

Li, A., Zhang, X., Lu, Y., et al. (2006). Cyclin D1 determines mitochondrial

function in vivo. Mol. Cell. Biol. 26, 5449–5469.

Sakaue-Sawano, A., Kurokawa, H., Morimura, T., Hanyu, A., Hama, H.,

Osawa, H., Kashiwagi, S., Fukami, K., Miyata, T., Miyoshi, H., et al. (2008).

Visualizing spatiotemporal dynamics of multicellular cell-cycle progression.

Cell 132, 487–498.

Scarpulla, R.C., Vega, R.B., and Kelly, D.P. (2012). Transcriptional integration

of mitochondrial biogenesis. Trends Endocrinol. Metab. 23, 459–466.

Schafer, E., Seelert, H., Reifschneider, N.H., Krause, F., Dencher, N.A., and

Vonck, J. (2006). Architecture of active mammalian respiratory chain super-

complexes. J. Biol. Chem. 281, 15370–15375.

Schieke, S.M., McCoy, J.P., Jr., and Finkel, T. (2008). Coordination of mito-

chondrial bioenergetics with G1 phase cell cycle progression. Cell Cycle 7,

1782–1787.

Sernagor, E. (2006). Retinal Development (Cambridge University Press).

Shimura, H., Hattori, N., Kubo, Si., Mizuno, Y., Asakawa, S., Minoshima, S.,

Shimizu, N., Iwai, K., Chiba, T., Tanaka, K., and Suzuki, T. (2000). Familial

Parkinson disease gene product, parkin, is a ubiquitin-protein ligase. Nat.

Genet. 25, 302–305.

Sicinski, P., Donaher, J.L., Parker, S.B., Li, T., Fazeli, A., Gardner, H., Haslam,

S.Z., Bronson, R.T., Elledge, S.J., and Weinberg, R.A. (1995). Cyclin D1 pro-

vides a link between development and oncogenesis in the retina and breast.

Cell 82, 621–630.

Taranova, O.V., Magness, S.T., Fagan, B.M., Wu, Y., Surzenko, N., Hutton,

S.R., and Pevny, L.H. (2006). SOX2 is a dose-dependent regulator of retinal

neural progenitor competence. Genes Dev. 20, 1187–1202.

Wang, C., Li, Z., Lu, Y., Du, R., Katiyar, S., Yang, J., Fu, M., Leader, J.E.,

Quong, A., Novikoff, P.M., and Pestell, R.G. (2006). Cyclin D1 repression of

nuclear respiratory factor 1 integrates nuclear DNA synthesis and mitochon-

drial function. Proc. Natl. Acad. Sci. USA 103, 11567–11572.

Wang, Z., Fan, M., Candas, D., Zhang, T.Q., Qin, L., Eldridge, A., Wachsmann-

Hogiu, S., Ahmed, K.M., Chromy, B.A., Nantajit, D., et al. (2014). Cyclin B1/

Cdk1 coordinates mitochondrial respiration for cell-cycle G2/M progression.

Dev. Cell 29, 217–232.

Wong-Riley, M. (1979). Changes in the visual system ofmonocularly sutured or

enucleated cats demonstrable with cytochrome oxidase histochemistry. Brain

Res. 171, 11–28.

Wu, Z., Puigserver, P., Andersson, U., Zhang, C., Adelmant, G., Mootha, V.,

Troy, A., Cinti, S., Lowell, B., Scarpulla, R.C., and Spiegelman, B.M. (1999).

Mechanisms controlling mitochondrial biogenesis and respiration through

the thermogenic coactivator PGC-1. Cell 98, 115–124.

Yaron, O., Farhy, C., Marquardt, T., Applebury, M., and Ashery-Padan, R.

(2006). Notch1 functions to suppress cone-photoreceptor fate specification

in the developing mouse retina. Development 133, 1367–1378.

Young, R.W. (1985). Cell differentiation in the retina of the mouse. Anat. Rec.

212, 199–205.

eports 14, 1684–1697, February 23, 2016 ª2016 The Authors 1697